Older adults are at increased risk for cognitive decline following major surgeries with this having implications for neurodegenerative acceleration as well as post-operative and long-term care costs. Orthopedic replacement surgeries such as total knee replacement surgery have the highest rates of cognitive decline in older adults. This is alarming, for older adults are increasingly seeking joint replacement to reduce associated osteoarthritis pain and increase activity (i.e., quality of life). Unfortunately at present there ae no specific surgical or anesthetic mechanisms for post-operative cognitive dysfunction. Results from our NIA funded pilot training grant, however, show preliminary evidence demonstrating that patients' pre-surgical brain integrity is an important indicator for post-operative outcome. We now seek to acquire definitive evidence for these markers using a larger sample size and also explore the interaction between pre-operative neuronal integrity and specific perioperative variables (e.g., anesthesia depth, emboli number). Aim 1 is to examine the integrity of the white matter regions connecting the frontal to subcortical structures and the hypothesis that patients with less deep white matter integrity will be vulnerable to post-operative executive decline. We further hypothesize that this disruption is driven by compromise to an important white matter circuit connecting the dorsolateral prefrontal cortex to the caudate (subaim 1). Aim 2 is to examine the integrity of the medial temporal structures to support the hypothesis that patients' with reduced pre-surgical hippocampal-entorhinal connectivity are vulnerable to post-operative memory decline. In Aim 3, we explore which peri-operative variables (e.g., intraoperative emboli, embolic stroke, anesthesia duration) interact negatively with baseline neuroimaging variables. These aims will be carried out by a multidisciplinary team with expertise in neuropsychology, geriatrics, dementia, post-operative cognitive dysfunction, anesthesiology, and orthopedic surgery. We will conduct a prospective longitudinal study with two groups: older adults (age > 60 years) having total knee replacement (n=80) and non-surgery age and education matched peers with osteoarthritis (n=80). Both groups will acquire baseline MRI using sophisticated diffusion and functional measures to define specific neuronal regions of interest, and complete cognitive testing at a pre-surgery/baseline time point followed by repeat testing at three weeks, three-months, and one-year post-operative/post-baseline. Participants will also acquire a repeat MRI 48 hours after their baseline scan to identify changes in brain structure (i.e., embolic stroke). Overall, the study's findings will bring us closer to understanding neural mechanisms for post- operative cognitive dysfunction, will help argue for basic pre-operative MRI screening for certain older adults, and assist us in the development of intra-operative interventions for patients with neuronal vulnerabilities.